The Most Advanced Implementation of the NCI Model

How many cups of coffee have you had today? How many pieces of fruit? Is this a realistic reflection of your usual or long-term consumption habits? Probably not. Even for the most regular coffee drinker, the amount they consumed may vary considerably from day to day.
The usual intake is the long-term intake of a food, nutrient or chemical and depends on both the frequency of consumption and the amounts consumed over an extended period of time.

A description of usual (long term) intakes at a population level characterises the variability that exists between individuals as well as the variability within the same person’s habits. As part of the new release of Creme Nutrition and Creme Food Safety applications, a new and improved modelling option is now available for the estimation of usual intakes of foods, nutrients and ingredients for populations around the world.

In the case of coffee, we know that consumption habits are likely to vary in any given population. Some people do not drink coffee at all, while others consume it occasionally, or even every day.

The NCI model, now available in the Creme Global applications, implements the non-linear mixed effects model devised by the National Cancer Institute to model the consumption of episodically consumed foods, and it allows users to easily include an estimate of usual intakes as part of any standard exposure assessment.

The Creme Global implementation of this model is the most, flexible and user-friendly representation of this model available anywhere. It makes it a straightforward task to analyse the usual intake patterns of any sub-population (from the US to China), any socio-economic sub-group(s), any food type (or basket of foods), and any additive, contaminant, or nutritional ingredient.

The NCI model uses multiple 24-hour recalls to infer usual intakes by explicitly including between-individual and within-individual variability, and Monte Carlo simulations produce empirical distributions of usual intakes.

When compared to the distribution of daily averages, the distribution of usual intakes offers a more reliable representation of long-term consumption, especially at the upper and lower percentiles. This is particularly relevant when considered in conjunction with reference values such as recommended daily intakes or guidance values to estimate what percentage of the population falls above or below.

When combined with our pre-installed and validated data sets, usual intakes of caffeine can be estimated on the basis of survey data for all of the major markets internationally.

If you would like information on habitual intakes of any food, ingredient or contaminant for any population around the world, you can contact us on

Further reading:
> Read more about our population predictive nutritional intake modelling application: Creme Nutrition.
> Read more about our food safety aggregate exposure modelling application: Creme Food Safety.


Written by Stefania Giammanco on January 22 2016

Signup for our newsletter